Why do I get the following error when I attempt to deploy my batch endpoint in Azure ML: "could not find datastore azureml"

Question

Why do I get the following error when I attempt to deploy my batch endpoint in Azure ML: "could not find datastore azureml"

Adam Goldammer 25

I very frequently get the following error when I attempt to deploy a batch endpoint: "could not find datastore azureml."

User's image

I am using Auto ML in Azure Machine Learning Studio. It doesn't happen all the time but it does happen frequently. I clearly have an "azureml" blob container in my storage account, so why would I be getting this error?

User's image

Manas Mohanty 5,950 Reputation points Microsoft External Staff Moderator

2025-06-04T17:30:16.9233333+00:00

Hi Adam Goldammer

Hope Amira Bedhiafi's pointers provided valuable insights to you.

Just wanted to add that your screenshot says azureml blob container is there, but please make sure you have kept workspace default store - workspaceblobstore as default and it is in correct state (not corrupt due to accidental deletion of metadata from container files or artifacts)

if not correct state, you can create another datastore from dedicated container and name it as azureml as Amira said

Thank you.

Accepted answer

0 additional answers

Your answer

Manas Mohanty 5,950 Reputation points Microsoft External Staff Moderator

2025-06-04T17:30:16.9233333+00:00

Hi Adam Goldammer

Hope Amira Bedhiafi's pointers provided valuable insights to you.

Just wanted to add that your screenshot says azureml blob container is there, but please make sure you have kept workspace default store - workspaceblobstore as default and it is in correct state (not corrupt due to accidental deletion of metadata from container files or artifacts)

if not correct state, you can create another datastore from dedicated container and name it as azureml as Amira said

Thank you.

Answer 1

Amira Bedhiafi 33,631 Volunteer Moderator

Hello Adam !

Thank you for posting on Microsoft Learn.

Azure ML is looking for a registered datastore named azureml, not just a blob container with that name. You do have a blob container called azureml in your storage account, but that alone doesn’t register it as a usable datastore in Azure ML.

If it's missing:

Go to Manage > Data > Datastores in Azure ML Studio
Click + New datastore
Register it using the following:
- Name: azureml
- Type: Azure Blob Storage
- Container name: azureml
- Point it to the correct Storage Account and Authentication method

If you don’t want to recreate a datastore named azureml:

Go to your batch deployment YAML or JSON config (if used)
Change the output and input storage references to use the name of a valid datastore.

Adam Goldammer 25 Reputation points

2025-06-05T14:14:29.3133333+00:00

This resolved the issue, Amira. Thank you so much for your help!
Amira Bedhiafi 33,631 Reputation points Volunteer Moderator

2025-06-06T12:08:14.02+00:00

You are always welcome ! ;D

Share via

Why do I get the following error when I attempt to deploy my batch endpoint in Azure ML: "could not find datastore azureml"

0 additional answers

Your answer